High mobility group box 1 potentiates thepro-inflammatory effects of interleukin-1b inosteoarthritic synoviocytesIsabel García-Arnandis1, Maria Isabel Guillén1,2, Francisco Gomar3, Jean-Pierre Pelletier4, Johanne Martel-Pelletier4,Maria José Alcaraz1*
Abstract
Introduction: High mobility group box 1 (HMGB1) is released by necrotic cells or secreted in response toinflammatory stimuli. Extracellular HMGB1 may act as a pro-inflammatory cytokine in rheumatoid arthritis. We haverecently reported that HMGB1 is released by osteoarthritic synoviocytes after activation with interleukin-1beta (IL-1b) The present study investigated the role of HMGB1 in synovial inflammation in osteoarthritis (OA).
Methods: HMGB1 was determined in human synovium using immunohistochemistry, comparing normal to OA.OA synoviocytes were incubated with HMGB1 at 15 or 25 ng/ml in the absence or presence of IL-1b (10 ng/ml).Gene expression was analyzed by quantitative PCR and protein expression by Western Blot and ELISA. Matrixmetalloproteinase (MMP) activity was studied by fluorometric procedures and nuclear factor (NF)-�B activation bytransient transfection with a NF-�B-luciferase plasmid.
Results: In the normal synovium, HMGB1 was found in the synovial lining cells, sublining cells, and in the vascularwall cells. The distribution of HMGB1 in OA synovium was similar but the number of HMGB1 positive cells washigher and HMGB1 was also present in infiltrated cells. In normal synovial membrane cells, HMGB1 was foundmostly in the nuclei, whereas in OA, HMGB1 was generally found mostly in the cytoplasm. In OA synoviocytes,HMGB1 alone at concentrations of 15 or 25 ng/ml did not affect the production of IL-6, IL-8, CCL2, CCL20, MMP-1or MMP-3, but in the presence of IL-1b, a significant potentiation of protein and mRNA expression, as well as MMPactivity was observed. HMGB1 also enhanced the phosphorylated ERK1/2 and p38 levels, with a lower effect onphosphorylated Akt. In contrast, JNK1/2 phosphorylation was not affected. In addition, HMGB1 at 25 ng/mlsignificantly potentiated NF-�B activation in the presence of IL-1b.Conclusions: Our results indicate that HMGB1 is overexpressed in OA synovium and mostly present in extracellularform. In OA synoviocytes, HMGB1 cooperates with IL-1b to amplify the inflammatory response leading to theproduction of a number of cytokines, chemokines and MMPs. Our data support a pro-inflammatory role for thisprotein contributing to synovitis and articular destruction in OA.
IntroductionThe nuclear DNA-binding protein high mobility groupbox 1 (HMGB1) can be passively released by necroticcells or secreted by macrophages and other myeloidcells in response to inflammatory stimuli as part of theinflammatory response to infection or injury (reviewed
in [1]). It is known that pro-inflammatory cytokinessuch as IL-1b or TNFa stimulate HMGB1 translocationinto the cytoplasm and release in different cell types,although TNFa is not the main inducer of extracellularHMGB1 during synovitis in rheumatoid arthritispatients [2]. Oxidative stress has also been shown toinduce HMGB1 release potentially through a mitogen-activated protein kinase (MAPK) and chromosomeregion maintenance mechanism [3].* Correspondence: [email protected]
1Department of Pharmacology, University of Valencia, Av.Vicent AndrésEstellés s/n, Burjasot, 46100 Valencia, SpainFull list of author information is available at the end of the article
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
There is abundant evidence that HMGB1 induces cellproliferation, migration and differentiation [4-6]. Bindingof HMGB1 to bacterial products might create complexesinducing innate immune responses and production ofinflammatory mediators. Extracellular HMGB1 interactswith receptor for advanced glycation end products(RAGE) and the toll-like receptors (TLR) includingTLR-2 and TLR-4 [1] leading to the activation of mono-cytes, macrophages and dendritic cells. In addition, theinteraction of HMGB1 with phosphatidylserine on thecell surface inhibits the phagocytosis of apoptotic neu-trophils by macrophages [7], which may retard the reso-lution of inflammation. Nevertheless, recent studiessuggest that HMGB1 alone demonstrates minor pro-inflammatory activity, which is potentiated throughbinding to IL-1b and other inflammatory mediators [8].HMGB1 plays a role as a pro-inflammatory cytokine
in rheumatoid arthritis and animal models of this dis-ease. Hence, HMGB1 is overexpressed in synovial tissueof rheumatoid arthritis patients and its extracellularform has been related to the progression of arthritis inanimal models [9]. Moreover, RAGE activation byHMGB1 results in increased invasiveness of fibroblast-like synoviocytes from rheumatoid arthritis patients [10].In human articular cartilage, HMGB1 may participate
in endochondral ossification during osteogenesis [11]and recently, the related protein HMGB2 has beeninvolved in ageing and osteoarthritis (OA) [12]. Studiesshow that HMGB1 [13] and its receptor RAGE [14] areexpressed in OA cartilage. It is also known that stimula-tion of OA chondrocytes with HMGB1 results in phos-phorylation of extracellular signal-regulated kinase(ERK) and nuclear factor-�B (NF-�B), and matrix metal-loproteinase (MMP) expression [14]. In addition, activa-tion of RAGE by advanced glycation end products(AGEs) in OA chondrocytes and synoviocytes leads toincreased catabolic activity and cartilage degeneration[15]. Nevertheless, the pro-inflammatory activity ofHMGB1 in synoviocytes and its participation in synovi-tis during OA remain to be determined. We haverecently reported that HMGB1 is released by OA syno-viocytes after activation with IL-1b [16], suggesting theparticipation of HMGB1 in the inflammatory responseinduced by this cytokine. In the present study, wefurther investigated the role of HMGB1 in OA synovialinflammation.
Materials and methodsSpecimen selectionHuman OA knee synovial membranes were obtainedfrom patients (12 female, 3 male, aged 69 ± 1 years,mean ± standard error of the mean (SEM)) undergoingtotal knee arthroplasty. All patients fulfilled the Ameri-can College of Rheumatology criteria for OA of the
knee [17]. Normal knees (2 female, 1 male, aged 72 ±1.5 years) were obtained within 12 hours of death; thetissues were examined macroscopically and microscopi-cally to ensure that only normal tissue was used. Thisstudy was approved by the Institutional Ethical Commit-tee and is in compliance with all ethical standards andpatients’ consent according to the Declaration ofHelsinki.
Immunohistochemistry of HMGB1 in human synoviumHMGB1 was determined in the human synovium usingimmunohistochemistry, comparing normal with OA.Tissues were processed following two methodologies.Firstly, we used an HMGB1 antibody that was revealedwith 3,3’-diaminobenzidine (DAB) for reading with lightmicroscope, and secondly a fluorescent HMGB1 anti-body for fluorescence microscopy analysis. In brief, thespecimens were dissected and fixed in TissuFix #2(Chaptec, Montreal, QC, Canada) and processed directlyafter acquisition from the donor for immunohistochem-istry (basal synthesis), as previously described [18]. Sec-tions of 5 μm of paraffin-embedded specimens weredeparaffinized in toluene and rehydrated in a gradedseries of ethanol. The synovium was treated with TritonX-100 (0.3%) for 20 minutes at room temperature.Slides were washed in PBS followed by 3% hydrogenperoxide/methanol for 15 minutes. They were furtherincubated for 45 minutes with 1.5% normal serum (Vec-tor Laboratories, Burlingame, CA, USA) and overlaidwith the rabbit anti-human HMGB1 antibody (1:50 dilu-tion; Abcam, Cambridge, MA, USA) for 18 hours at 4°Cin a humidified chamber. The second antibody was agoat anti-rabbit immunoglobulin (Vector Laboratories,Burlingame, CA, USA) in which slides were incubatedfor 45 minutes. For reading with a light microscope,slides were stained using the avidin-biotin complexmethod (Vectastain ABC kit, Vector Laboratories, Bur-lingame, CA, USA). The colour was developed withDAB (DAKO Diagnostics Canada Inc., Mississauga, ON,Canada) without nickel chloride. The slides were coun-ter-stained with Mayer’s hematoxylin. Each section wasexamined under a light microscope (Leitz Orthoplan;Leica Inc., St-Laurent, QC, Canada) and photographedwith a Retega OEM Fast camera (QImaging, Surrey, BC,Canada). For detection with fluorescence of Alexa 488fluorochrome, slides were stained with TSA™ #22 (Invi-trogen, Burlington, ON, Canada) and mounted withVectashield (Vector Laboratories, Burlingame, CA, USA)containing 4’,6-diamidino-2-phenylindole (DAPI) inorder to stain the nucleus. The slides were examinedunder a fluorescence microscope and photographed by aCoolSNAP cf Photometrics camera (Roper Scientific,Rochester, NY, USA). Controls were performed to deter-mine the specificity of staining: i) substitution of the
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
Page 2 of 12
primary antibodies with non-immune isotype rabbit IgG(used at the same concentration as the primary anti-body); ii) immunoadsorbed with 20-fold excess of thepeptide HMGB1; and iii) omission of primary antibody.Controls showed only background staining.The presence of the antigen in the synovium was
quantified by determining the number of cells thatstained positive with the distinction between nuclearand extra-nuclear (cytoplasmic and cell vicinity) stain-ing. Each synovial membrane was divided into threemicroscopic fields (40×; Leitz Orthoplan) and the totalnumber of cells and those staining positive for the speci-fic antigen were determined in the lining cells for bothnormal and OA, and in the infiltrates for OA specimens.The final results were expressed as the percentage ofcells staining positive for the antigen with the maximumscore being 100%.Moreover, the synovial lining cell hyperplasia was
graded on a scale of 0 to 2 as previously described [19],where 0 = 1 to 2 layers of cells, 1 = 3 to 5 layers, and 2= 6 or more layers. The layers were counted from thesurface of the membrane to the subsynovial tissue. Eachsynovium was graded into three microscopic fields (40×;Leitz Orthoplan) and the score averaged.
Confocal microscopyDiscrimination of the HMGB1 location within the cellswas confirmed using immunofluorescence followed byconfocal microscopy. In brief, sections of 10 μm of par-affin-embedded specimens were treated as above. Con-focal acquisitions were carried out with a Leica TCSSP5 broadband confocal microscope (St-Laurent, QC,Canada, which uses confocal point-scanning for opticalsectioning. In addition, it has a spectral imaging detectorallowing tunable emission bands and dye separation.The system is equipped with the AOBS (Acousto-Opti-cal Beam Splitter) for optical beam splitting. It has threelasers: Ar 458/488/514, DPSS 561 and HeNe 633, 543,594.
Cell culture and treatmentsSynovial specimens were finely minced and isolated byenzymatic digestion with collagenase type 1A (SigmaAldrich, St Louis, MO, USA) in DMEM/HAM F12(Sigma-Aldrich, St Louis, MO, USA) containing penicil-lin (100 U/ml) and streptomycin (100 μg/ml) at 37°C in5% carbon dioxide atmosphere for 16 hours. Thedigested tissue was filtered through a 70 mm nylonmesh, washed and centrifuged. Cell viability was greaterthan 95% according to the Trypan blue exclusion test.Collected cells were resuspended in DMEM/HAM F12(Sigma-Aldrich, St Louis, MO, USA) containing penicil-lin (100 U/ml) and streptomycin (100 μg/ml) supple-mented with 10% fetal bovine serum (Sigma-Aldrich, St
Louis, MO, USA) and cultured at 37°C in 5% carbondioxide atmosphere until third passage (95% fibroblasts,detected by immunocytochemistry with anti-collagen Iantibody (Chemicon, Millipore Iberica, Madrid, Spain)).Synoviocytes were allowed to grow to nearly confluenceand incubated with HMGB1 (HMGBiotech, Milano,Italy) at 15 or 25 ng/ml and IL-1b (10 ng/ml, PeprotechEC Ltd, London, UK) or culture medium. Viability stu-dies were performed for all the experimental conditions.None of the treatments significantly affected cell viabi-lity which was more than 90% as tested by Trypan blueexclusion. The possibility of endotoxin contamination ofHMGB1 was excluded after performing the Limulus testusing the commercial kit from Sigma-Aldrich (St Louis,MO, USA; data not shown).
Western blot analysisAfter stimulation for five minutes with IL-1b (10 ng/ml), HMGB1 at 15 and 25 ng/ml or IL-1b+HMGB1,synoviocytes were lysed in 100 μl of buffer (1% TritonX-100, 1% deoxycholic acid, 20 mM NaCl and 25 mMTris, pH 7.4) and centrifuged at 4°C for 15 minutes at10,000g. Proteins (25 μg) in cell lysates were separatedby 12.5% SDS-PAGE and transferred onto polyvinyli-dene difluoride membranes. Membranes were blockedwith 3% bovine serum albumin and incubated withspecific antibodies against phosphorylated ERK (CellSignaling Technology, Inc., Beverly, MA, USA, dilu-tion 1:800), phosphorylated or total Akt and ERK(Cell Signaling Technology, Inc., Beverly, MA, USA;dilution 1:500), and phosphorylated or total c-Jun N-terminal kinase (JNK; Cell Signaling Technology, Inc.,Beverly, MA, USA; dilution 1:250) and p38 (PromegaCorporation, Madison, WI, USA; dilution 1:250) over-night at 4°C. Finally, membranes were incubated withperoxidase-conjugated goat anti-rabbit IgG (Dako,Copenhagen, Denmark) and the immunoreactivebands were visualized by enhanced chemiluminescence(GE Healthcare, city, Barcelona, Spain) using theAutoChemi image analyzer (UVP Inc., Upland, CA,USA).
Determination of MMP activityCells were stimulated with IL-1b (10 ng/ml), HMGB1 at15 and 25 ng/ml or IL-1b+HMGB1 for 24 hours andsupernatants were harvested, centrifuged and incubatedwith p-aminophenyl mercuric acetate for six hours at37°C to activate MMPs. Aliquots of supernatants werethen transferred to a 96-well plate and after addition ofthe 5-FAM peptide substrate (AnaSpec Inc., San Jose,CA, USA), fluorescence was measured for differenttimes at 490 nm (excitation)/520 nm (emission) in aVictor3 microplate reader (PerkinElmer España, Madrid,Spain).
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
Page 3 of 12
Enzyme-linked immunosorbent assaySynoviocytes were stimulated with IL-1b (10 ng/ml) for24 hours, in the presence or absence of HMGB1 at 15and 25 ng/ml. Supernatants were harvested, centrifugedand frozen at -80°C until analysis. IL-6, IL-8 and chemo-kine (C-C motif) ligand 2 (CCL2) levels were deter-mined by specific ELISA from eBioscience (San Diego,CA, USA) with sensitivity of 2, 4 and 7 pg/ml, respec-tively. CCL20 was determined with a specific ELISAfrom Raybiotech Inc. (Norcross, GA, USA) with sensitiv-ity of 1.5 pg/ml. MMP-1 and MMP-13 protein wasquantified in supernatants by using specific ELISAsfrom AnaSpec (sensitivity of 8 and 6 pg/ml, respec-tively), and MMP-3 with the ELISA from RaybiotechInc. (sensitivity of 0.3 ng/ml).
Real-time PCRFollowing incubation for 24 hours, total RNA wasextracted using the TriPure reagent (Roche AppliedScience, Barcelona, Spain) according to the manufac-turer’s instructions. Reverse transcription was accom-plished on 1 μg of total RNA using random primers(TaqMan reverse transcription reagents, Applied Biosys-tems, Spain, Madrid). PCR assays were performed induplicate on an iCycler Real-Time PCR Detection Sys-tem using SYBR Green PCR Master Mix (Bio-RadLaboratories, Richmond, CA, USA) [20]. Sequences ofprimers used have been reported previously [21-24]. Foreach sample, differences in threshold cycle (DCt) valueswere calculated by correcting the Ct of the gene ofinterest to the Ct of the reference gene glyceraldehide-3-phosphate dehydrogenase. Relative gene expressionwas expressed as DDCt with respect to nonstimulatedcells.
Activation of NF-�BCells were seeded into six-well plates and grown to 50to 60% confluence. Transient transfection was per-formed for 45 minutes with 2 μg of the reporter con-struct NF-�B-luc (Stratagene, La Jolla, CA, USA) and 1μg of the internal control pRL-TK (Promega Corpora-tion, Madison, WI, USA) by the Magnetofection™ sys-tem (OZ Biosciences, Marseille, France) according tothe manufacturer’s recommendations. The mediumwas then replaced and cells were treated for 24 hourswith HMGB1 at 25 ng/ml in the absence or presenceof IL-1b (10 ng/ml). After lysis and centrifugation, ali-quots of supernatants were used to assay firefly andRenilla luciferase activity using the Dual-LuciferaseReporter Assay System kit (Promega Corporation,Madison, WI, USA). Luminescence was measured in aMicrobeta counter (Wallac, Turku, Finland) and fireflyluciferase activity was normalized to Renilla luciferaseactivity.
Statistical analysisResults are presented as mean ± SEM. Statistical ana-lyses were performed using one-way analysis of variancefollowed by Dunnett’s t-test for multiple comparisonsand two-tailed unpaired Student’s t-test for dualcomparisons.
ResultsHMGB1 expression in human synoviumIn the normal and OA synovium, HMGB1 was found inthe cells of the synovial lining, sublining cells, and inthe cells of the vascular wall (i.e. cells around the bloodvessels) (Figure 1a). We further performed an analysis ofthe percentage of positive synoviocytes in the nuclearand extra-nuclear (cytoplasmic and cell vicinity) com-partments of the cells comparing normal (n = 3) withOA (n = 3). As illustrated in Figure 1b, a statisticallysignificant decrease in HMGB1 was found in OA cellnuclei when compared with normal (P < 0.03), and,although there was more positive staining in the extra-nuclear (cytoplasm and in the cell vicinity) compart-ment, this did not quite reach statistical significance (P< 0.07). In OA synovium, the percentage of infiltratecells staining positive in the nuclei (23.6 ± 3.9%) wassignificantly less than in the extra-nuclear compartment(52.4 ± 4.1%; P < 0.007). Moreover, comparison of thesynovial lining layers showed, as expected, more celllayers in the OA (score:1.3 ± 0.2) than in the normal(0.6 ± 0.2) synovial membrane.Experiments performed with the fluorescent antibody
(Figure 2) confirmed that in normal synovial membranecells, HMGB1 (green) was found mostly in the nucleiwith some cells showing cytoplasm staining (Figures 2ato 2c). In OA (Figures 2b to 2d), however, HMGB1 wasgenerally found mostly in the cytoplasm. For cell infil-trates (Figures 2e to 2f), the pattern was as for the OAlining cells, in which HMGB1 was found mostly in thecytoplasm.
Effect of HMGB1 on cytokine and chemokine productioninduced by IL-1b in OA synoviocytesTo determine whether extracellular HMGB1 can modu-late cytokine and chemokine production in human OAsynoviocytes, cells were incubated with HMGB1 in thepresence or absence of IL-1b (10 ng/ml). This pro-inflammatory cytokine is present in OA synovial fluidand cartilage and participates in joint degradation[25,26]. As expected, stimulation with IL-1b resulted inthe enhanced production of pro-inflammatory cytokinesand chemokines. As shown in Figure (3a to 3d),although HMGB1 alone at concentrations of 15 or 25ng/ml did not affect the production of IL-6, IL-8, CCL2or CCL20, a significant potentiation of these pro-inflam-matory mediators was observed in the presence of IL-
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
Page 4 of 12
1b. These effects of HMGB1 were confirmed at themRNA level, with an enhancement of IL-6, IL-8, CCL2and CCL20 mRNA expression in cells stimulated withIL-1b (Figures 4a and 4b).
Effect of HMGB1 on MMPs induced by IL-1b in OAsynoviocytesCell activation by IL-1b (10 ng/ml) potently inducedMMP gene expression as well as MMP protein andactivity. mRNA expression was measured by real-timePCR. In cell supernatants, protein levels were measuredby ELISA and MMP activity by a fluorometric proce-dure. As shown in Figure (5a to 5c), HMGB1 alone didnot induce significant changes in MMP-1, MMP-3 orMMP-13 protein in cell supernatants, but it potentiatedthe stimulating effect of IL-1b for MMP-1 and MMP-3.In addition, HMGB1 significantly increased MMP-1 andMMP-3 mRNA expression in the presence of IL-1b(Figure 6a). These effects were reflected in the levels of
MMP activity released into the medium, which were sig-nificantly increased by HMGB1 after IL-1b stimulation(Figure 6b).
Effect of HMGB1 on Akt and MAPK phosphorylationinduced by IL-1b in OA synoviocytesTo determine the possible mechanism of action ofHMGB1, we further examined whether this proteinacts on Akt and MAPK activation. The time of stimu-lation for a high phosphorylation response was chosenfrom previous experiments [see Additional file 1]. Asshown in Figure 7, HMGB1 at both concentrationsstudied (15 and 25 ng/ml) increased the phosphory-lated Akt and ERK1/2 levels. In the presence of IL-1b,HMGB1 enhanced the phosphorylated ERK1/2 andp38 levels, with a lower effect on phosphorylated Akt.In contrast, JNK1/2 phosphorylation was not affectedby HMGB1 either in the presence or absence of IL-1bstimulation.
Figure 1 Expression of HMGB1 protein in human synovium. (a) Representative immunohistochemical sections of normal (n = 3) andosteoarthritic (n = 3) human synovium with high mobility group box 1 (HMGB1) revealed in 3,3’-diaminobenzidine (DAB). Arrows indicateHMGB1 positive cells. Original magnification × 100. (b) Histograms of normal and osteoarthritic positive cells (%) in the synovial lining cellnucleus and cytoplasm and vicinity. Data are expressed as mean ± standard error of the mean (n = 3). *P < 0.05 with respect to normal.
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
Page 5 of 12
Effect of HMGB1 on NF-�B activation induced by IL-1b inOA synoviocytesNF-�B is a main regulator of pro-inflammatory anddegradative genes in the joint. As IL-1b [27] andHMGB1 [28] cell stimulation results in the activation ofthis pathway in different cell types, we studied the possi-ble participation of this mechanism in the pro-inflam-matory effects of HMGB1 in OA synoviocytes. HMGB1treatment in nonstimulated cells resulted in increasedtranscriptional activity, although it did not reach statisti-cal significance. Of note, HMGB1 at 25 ng/ml signifi-cantly potentiated NF-�B activation in the presence ofIL-1b (Figure 8).
DiscussionSynovial inflammation has been demonstrated in tissuesamples of OA patients and may be related to diseaseprogression [17,29,30]. Several lines of evidence supportthe involvement of synoviocytes in OA cartilage degra-dation through the production of inflammatory andcatabolic mediators [31]. In this regard, pro-inflamma-tory cytokines such as IL-1b play a role in driving syno-vitis during OA and influencing the production ofcytokines and MMPs [32]. With respect to the biologicalrole of HMGB1, it is essential to understand themechanisms involved in the regulation of the inflamma-tory response. HMGB1 appears to act as a pro-inflam-matory cytokine in mononuclear cells through the
release of TNFa, IL-1b, IL-6, IL-8, macrophage inflam-matory protein-1 and nitric oxide [4,33]. In addition,HMGB1 stimulates the motility of a wide range of cellsand thus elicits the migration to the site of tissuedamage, a process dependent on the activation of ERK,NF-�B [34] and Src [35] pathways.Our data first showed that less HMGB1 is observed
in the nuclei of OA synoviocytes, and although ahigher level is found in the cytoplasm and vicinity inthese cells, this did not quite reach statistical signifi-cance. This lack of significance could be related to thelow number of specimens analysed; however, asHMGB1 is released extra-cellularly, our evaluationcould be falsely lower. Yet, as there were more celllayers in OA compared with normal, in addition to thepresence of cell infiltrates in OA showing an elevatedlevel of HMGB1 in the extra-nuclear compartmentcompared with the nuclear, this strongly suggests thatmore HMGB1 is released from OA synovial cells thanfrom normal, supporting the previous data in whichOA cells released HMGB1 upon activation by IL-1b[16]. We thus hypothesized that HMGB1 may play arole in modulating the inflammatory process in syno-vium during OA. Our results indicate that HMGB1cooperates with IL-1b to amplify the inflammatoryresponse leading to the production of a number ofcytokines, chemokines and MMPs in OA synoviocytes.Data showed that HMGB1 plus IL-1b synergistically
Figure 2 Confocal analysis of HMGB1 expression in human synovium. Representative immunohistochemical sections of human (a, c)normal (n = 3) and (b, d) osteoarthritic (n = 3) synovium with high mobility group box 1 (HMGB1) revealed in fluorescence and read withconfocal microscopy. (e) represents immunohistochemistry revealed in 3,3’-diaminobenzidine (DAB) of human osteoarthritic synovium infiltrates,and (f) is as (e) but revealed with the fluorescent antibody. The colour green represents HMGB1 and blue the nucleus. Original magnification ×100. (c) and (d) were amplified by the image processor software.
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
Page 6 of 12
enhanced IL-6 production, which is in line with arecent report on the stimulation of IL-6 release by pre-formed complexes of HMGB1 (100 ng/ml) and IL-1bat a low concentration unable to stimulate synovio-cytes (0.05 ng/ml). Interestingly, the response of syno-vial fibroblasts from OA or rheumatoid arthritispatients was similar [36].Chemokines such as IL-8, CCL2 and CCL20 are pro-
duced by OA synoviocytes, but to a lower extent thanrheumatoid arthritis cells. These mediators have the abil-ity to attract inflammatory cells and regulate gene
transcription and cell proliferation [37]. The up-regula-tion of chemokines upon IL-1b synoviocyte stimulationpromotes inflammation and cartilage degradationthrough the activation of MMPs and other degradativeenzymes [38]. CCL2 and CCL20 are chemokines impli-cated in rheumatoid arthritis synovitis [39-41] and areproduced by OA synovium in the presence of pro-inflam-matory cytokines. In particular, IL-1b has been shown tobe a more potent inducer of CCL20 than TNFa or IL-17[42]. Our data revealed that HMGB1 acted on OA syno-viocytes in vitro to enhance the production of IL-8, CCL2and CCL20. The best studied CXC chemokine is IL-8,which is produced by fibroblasts and macrophages pre-sent in synovial tissues [43]. Induction of synovial macro-phage and fibroblast chemotaxis has been demonstratedfor CCL2 [40,44], whereas CCL20 induces monocyte and
Figure 3 Effect of HMGB1 and IL-1b on the levels of cytokine andchemokine released into the medium by osteoarthriticsynoviocytes. (a) IL-6, (b) IL-8, (c) CCL2 and (d) CCL20 protein levels.Cells were stimulated with IL-1b (10 ng/ml) for 24 hours in thepresence or absence of high mobility group box 1 (HMGB1) at 15 and25 ng/ml. Protein levels were determined in supernatants by ELISA.Data are expressed as mean ± standard error of the mean. Duplicatesamples from six patients were used. ++P < 0.01 with respect tononstimulated cells. *P < 0.05, **P < 0.01 with respect to IL-1b.
Figure 4 Effect of HMGB1 and IL-1b on cytokine andchemokine mRNA levels in OA synoviocytes. (a) IL-6, IL-8, and(b) CCL2 and CCL20 mRNA relative expression. Cells werestimulated with IL-1b (10 ng/ml) for 24 hours in the presence orabsence of high mobility group box 1 (HMGB1) at 15 and 25 ng/ml.mRNA expression was determined by real-time PCR. Data areexpressed as mean ± standard error of the mean. Duplicate samplesfrom four patients were used. ++P < 0.01 with respect tononstimulated cells. *P < 0.05 with respect to IL-1b.
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
Page 7 of 12
memory lymphocyte chemotaxis from peripheral bloodto the rheumatoid joint [45]. Therefore, the enhancedproduction of these mediators in the presence ofHMGB1 supports a role for this protein in the amplifica-tion of the inflammatory response induced by IL-1b inOA synoviocytes.MMPs play an important role in articular tissue degra-
dation in OA. The present study demonstrated the poten-tiating effect of HMGB1 on IL-1b induction of MMP-1and MMP-3 in human OA synoviocytes. As MMP-1degrades collagens in the extracellular matrix [46-48] andMMP-3 activity leads to activation of collagenases [49],
our results suggest the amplification of catabolic responsesby HMGB1 during joint inflammation.Several studies indicate that spontaneous or stimulated
production of many inflammatory and degradative med-iators by OA synoviocytes are related to NF-�B activa-tion [50,51]. In particular, transcription of IL-6 [52],IL-8 [53], CCL2, CCL20 [41,54] and MMPs [55,56] isNF-�B dependent. Our data show that IL-1b plusHMGB1 synergistically increased the transcriptionalactivity of NF-�B, leading to an enhanced production ofinflammatory and catabolic mediators.MAPK activity regulates the activation of transcription
factors relevant in inflammatory responses [57]. We
Figure 5 Effect of HMGB1 and IL-1b on the levels of MMPreleased into the medium by osteoarthritic synoviocytes. (a)Matrix metalloproteinase (MMP)-1, (b) MMP-3 and (c) MMP-13protein levels in the medium. Cells were stimulated with IL-1b (10ng/ml) for 24 hours in the presence or absence of high mobilitygroup box 1 (HMGB1) at 15 and 25 ng/ml. MMP protein wasmeasured by ELISA in supernatants. Data are expressed as mean ±standard error of the mean. Duplicate samples from six to eightpatients were used. ++P < 0.01 with respect to nonstimulated cells.*P < 0.05, **P < 0.01 with respect to IL-1b.
Figure 6 Effect of HMGB1 and IL-1b on MMP mRNA levels inosteoarthritic synoviocytes and MMP activity released into themedium. (a) Matrix metalloproteinase (MMP)-1, MMP-3 and MMP-13mRNA levels and (b) MMP activity. Cells were stimulated with IL-1b(10 ng/ml) for 24 hours in the presence or absence of high mobilitygroup box 1 (HMGB1) at 15 and 25 ng/m. mRNA was determinedby real-time PCR and MMP activity was measured in supernatantsby a fluorometric procedure, as indicated in Materials and Methods.Data are expressed as mean ± standard error of the mean.Duplicate samples from four patients were used. ++P < 0.01 withrespect to nonstimulated cells. *P < 0.05, **P < 0.01 with respect toIL-1b.
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
Page 8 of 12
have shown that HMGB1 enhances p38 phosphoryla-tion, which participates in IL-6 and IL-8 transcription inhuman fibroblast-like synoviocytes [58,59]. As the pro-duction of MMP-1 and MMP-3 upon stimulation ofsynoviocytes with IL-1b depends on ERK activation [60],
we investigated the effects of HMGB1 on ERK phos-phorylation. Our results indicate that HMGB1 potenti-ates the effects of IL-1b on ERK phosphorylation, whichmay play a role in the up-regulation of MMP-1 andMMP-3 by HMGB1. In contrast, we did not observe anymodification of JNK1/2, which phosphorylates c-Jun andregulates the transcription factor activating protein-1[61] as well as erythroblastosis 26 transcription factors[62]. Interestingly, HMGB1 potentiates Akt phosphory-lation by IL-1b, a pathway involved in cell survival andproliferation of fibroblasts in rheumatoid arthritis syno-vium [63]. In addition, Akt activation may play a role inhuman cartilage breakdown, because it has been impli-cated in MMP-13 and aggrecanase-1 expression inducedby oncostatin M [64] and the synergistic induction ofMMP-1 and MMP-13 expression after oncostatin M+IL-1b stimulation of human chondrocytes [65]. Ourdata would therefore suggest that the potentiation ofERK, p38 and Akt activation by HMGB1 may be amechanism relevant for the increase in the intensity andpersistence of synovitis as well as the expression of cata-bolic factors in OA.
ConclusionsHMGB1 is overexpressed in synovial membranes of OApatients. The results presented here support the view thatHMGB1 would act as a pro-inflammatory cytokine, whichenhances the OA synovial inflammatory process. HMGB1was found to synergize with IL-1b to induce phosphoryla-tion of ERK1/2, p38 and Akt, as well as NF-�B activation.These effects result in the production of pro-inflammatoryand catabolic mediators that would contribute to synovitisand articular destruction during OA.
Figure 7 Effect of HMGB1 and IL-1b on Akt and MAPKphosphorylation. Cells were stimulated with IL-1b (10 ng/ml) forfive minutes in the presence or absence of high mobility group box1 (HMGB1) at 15 and 25 ng/ml. Protein level was determined in celllysates by western blotting by using specific antibodies againstphosphorylated or total proteins. Relative expression ofphosphorylated and total protein bands was calculated afterdensitometric analysis. Data are expressed as mean ± standard errorof the mean (samples from three patients). ++P < 0.01 with respectto nonstimulated cells. *P < 0.05, with respect to IL-1b. ERK,extracellular signal-regulated kinase; JNK, c-Jun N-terminal kinase;MAPK, mitogen-activated protein kinase.
Figure 8 Effect of HMGB1 and IL-1b on NF-�B activation .Transient transfection was performed with the reporter constructnuclear factor (NF)-�B-luc and the internal control pRL-TK, asindicated in Materials and Methods. Cells were treated for 24 hourswith high mobility group box 1 (HMGB1) at 25 ng/ml in theabsence or presence of IL-1b (10 ng/ml). Firefly luciferase activitywas normalized to Renilla luciferase activity. Data are expressed asmean ± standard error of the mean. Duplicate samples from sixpatients were used. ++P < 0.01 with respect to nonstimulated cells;*P < 0.05 with respect to IL-1b.
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
Page 9 of 12
Additional material
Additional file 1: Supplementary figure S1. Time course of highmobility group box 1 (HMGB1) and IL-1b effects on Akt and mitogen-activated protein kinase (MAPK) phosphorylation.
AbbreviationsAGE: advanced glycation end product; CCL: chemokine (C-C motif) ligand;DAB: 3,3’-diaminobenzidine; DAPI: 4’,6-diamidino-2-phenylindole; ELISA:enzyme-linked immunosorbent assay; ERK: extracellular signal-regulatedkinase; HMGB1: high mobility group box 1; IL: interleukin; JNK: c-Jun N-terminal kinase; MAPK: mitogen-activated protein kinase; MMP: matrixmetalloproteinase; NF-�B: nuclear factor-�B; OA: ostearthritis; PBS:phosphate-buffered saline; PCR: polymerase chain reaction; RAGE: receptorfor advanced glycation end products; SEM: standard error of the mean; TLR:toll-like receptor; TNFa: tumor necrosis factor-a.
AcknowledgementsThe authors are grateful to Martin Boily and Christelle Boileau, PhD, from theOsteoarthritis Research Unit, University of Montreal Hospital Research Centre(CRCHUM, Montréal, Québec, Canada), for their expertise inimmunohistochemistry and confocal microscopy. This work was supportedby grants SAF2007-61769 and RETICEF RD06/0013/2001 (Spanish Ministery ofScience and Innovation+FEDER). I. García-Arnandis thanks GeneralitatValenciana for a fellowship.
Author details1Department of Pharmacology, University of Valencia, Av.Vicent AndrésEstellés s/n, Burjasot, 46100 Valencia, Spain. 2Department of Chemistry,Biochemistry and Molecular Biology, Cardenal Herrera-CEU University, Av.Seminario s/n, Moncada, 46113 Valencia, Spain. 3Department of Surgery,School of Medicine, University of Valencia, Av. Blasco Ibáñez 15, Valencia,46010 Valencia, Spain. 4Osteoarthritis Research Unit, University of MontrealHospital Research Centre (CRCHUM), Notre-Dame Hospital, Montreal, 1560Rue Sherbrooke East, Montreal, Quebec H2L 4M1, Canada.
Authors’ contributionsIG and MIG carried out the experimental work and the data collection andinterpretation. FG participated in the design and coordination ofexperimental work, and acquisition of data. J-PP and JM-P participated inthe study design, data collection, analysis of data and preparation of themanuscript. MJA carried out the study design, the analysis and interpretationof data and drafted the manuscript. All authors read and approved the finalmanuscript.
Competing interestsThe authors declare that they have no competing interests.
Received: 26 January 2010 Revised: 9 July 2010Accepted: 27 August 2010 Published: 27 August 2010
References1. Pisetsky DS, Erlandsson-Harris H, Andersson U: High-mobility group box
protein 1 (HMGB1): an alarmin mediating the pathogenesis of rheumaticdisease. Arthritis Res Ther 2008, 10:209.
2. Sundberg E, Grundtman C, af Klint E, Lindberg J, Ernestam S, Ulfgren AK,Harris HE, Andersson U: Systemic TNF blockade does not modulatesynovial expression of the pro-inflammatory mediator HMGB1 inrheumatoid arthritis patients - a prospective clinical study. Arthritis ResTher 2008, 10:R33.
3. Tang D, Shi Y, Kang R, Li T, Xiao W, Wang H, Xiao X: Hydrogen peroxidestimulates macrophages and monocytes to actively release HMGB1. JLeukoc Biol 2007, 81:741-747.
4. Andersson U, Wang H, Palmblad K, Aveberger AC, Bloom O, Erlandsson-Harris H, Janson A, Kokkola R, Zhang M, Yang H, Tracey KJ: High mobilitygroup 1 protein (HMG-1) stimulates proinflammatory cytokine synthesisin human monocytes. J Exp Med 2000, 192:565-570.
5. Dumitriu IE, Bianchi ME, Bacci M, Manfredi AA, Rovere-Querini P: Thesecretion of HMGB1 is required for the migration of maturing dendriticcells. J Leukoc Biol 2007, 81:84-91.
6. Dumitriu IE, Baruah P, Valentinis B, Voll RE, Herrmann M, Nawroth PP,Arnold B, Bianchi ME, Manfredi AA, Rovere-Querini P: Release of highmobility group box 1 by dendritic cells controls T cell activation via thereceptor for advanced glycation end products. J Immunol 2005,174:7506-7515.
7. Liu G, Wang J, Park YJ, Tsuruta Y, Lorne EF, Zhao X, Abraham E: Highmobility group protein-1 inhibits phagocytosis of apoptotic neutrophilsthrough binding to phosphatidylserine. J Immunol 2008, 181:4240-4246.
8. Sha Y, Zmijewski J, Xu Z, Abraham E: HMGB1 develops enhancedproinflammatory activity by binding to cytokines. J Immunol 2008,180:2531-2537.
9. Andersson U, Erlandsson-Harris H: HMGB1 is a potent trigger of arthritis. JIntern Med 2004, 255:344-350.
10. Steenvoorden MM, Toes RE, Ronday HK, Huizinga TW, DeGroot J: RAGEactivation induces invasiveness of RA fibroblast-like synoviocytes invitro. Clin Exp Rheumatol 2007, 25:740-742.
11. Taniguchi N, Yoshida K, Ito T, Tsuda M, Mishima Y, Furumatsu T, Ronfani L,Abeyama K, Kawahara K, Komiya S, Maruyama I, Lotz M, Bianchi ME,Asahara H: Stage-specific secretion of HMGB1 in cartilage regulatesendochondral ossification. Mol Cell Biol 2007, 27:5650-5663.
12. Taniguchi N, Carames B, Ronfani L, Ulmer U, Komiya S, Bianchi ME, Lotz M:Aging-related loss of the chromatin protein HMGB2 in articular cartilageis linked to reduced cellularity and osteoarthritis. Proc Natl Acad Sci USA2009, 106:1181-1186.
13. Attur M, Dave M, Akamatsu M, Nakagawa N, Miki J, Yang H, Katoh M,Wisniewski J, Tracey K, Amin A: Differential expression of high mobilitygroup protein in human normal and arthritic cartilage; Functionalgenomic analysis. Trans Orthop Res So 2003, 28:18.
14. Loeser RF, Yammani RR, Carlson CS, Chen H, Cole A, Im HJ, Bursch LS,Yan SD: Articular chondrocytes express the receptor for advancedglycation end products: Potential role in osteoarthritis. Arthritis Rheum2005, 52:2376-2385.
15. Steenvoorden MM, Huizinga TW, Verzijl N, Bank RA, Ronday HK, Luning HA,Lafeber FP, Toes RE, DeGroot J: Activation of receptor for advancedglycation end products in osteoarthritis leads to increased stimulation ofchondrocytes and synoviocytes. Arthritis Rheum 2006, 54:253-263.
16. Garcia-Arnandis I, Guillen MI, Castejon MA, Gomar F, Alcaraz MJ: Hemeoxygenase-1 down-regulates high mobility group box 1 and matrixmetalloproteinases in osteoarthritic synoviocytes. Rheumatology 2010,49:854-861.
17. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, Christy W,Cooke TD, Greenwald R, Hochberg M: Development of criteria for theclassification and reporting of osteoarthritis. Classification ofosteoarthritis of the knee. Diagnostic and Therapeutic CriteriaCommittee of the American Rheumatism Association. Arthritis Rheum1986, 29:1039-1049.
18. Monfort J, Tardif G, Roughley P, Reboul P, Boileau C, Bishop PN, Pelletier JP,Martel-Pelletier J: Identification of opticin, a member of the small leucine-rich repeat proteoglycan family, in human articular tissues: a noveltarget for MMP-13 in osteoarthritis. Osteoarthritis Cartilage 2008,16:749-755.
19. Pelletier JP, Martel-Pelletier J, Ghandur-Mnaymneh L, Howell DS,Woessner JF Jr: Role of synovial membrane inflammation in cartilagematrix breakdown in the Pond-Nuki dog model of osteoarthritis. ArthritisRheum 1985, 28:554-561.
21. Mietz H, Esser JM, Welsandt G, Kociok N, Hueber A, Joussen A, Esser P,Krieglstein GK: Latanoprost stimulates secretion of matrixmetalloproteinases in tenon fibroblasts both in vitro and in vivo. InvestOphthalmol Vis Sci 2003, 44:5182-5188.
22. Boileau C, Pelletier JP, Tardif G, Fahmi H, Laufer S, Lavigne M, Martel-Pelletier J: The regulation of human MMP-13 by licofelone, an inhibitorof cyclooxygenases and 5-lipoxygenase, in human osteoarthriticchondrocytes is mediated by the inhibition of the p38 map kinasesignaling pathway. Ann Rheum Dis 2005, 64:891-898.
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
23. Kwon JH, Keates S, Bassani L, Mayer LF, Keates AC: Colonic epithelial cellsare a major site of macrophage inflammatory protein 3alpha (MIP-3alpha) production in normal colon and inflammatory bowel disease.Gut 2002, 51:818-826.
24. Lockwood CJ, Paidas M, Krikun G, Koopman L, Masch R, Kuczynski E,Kliman H, Baergen RN, Schatz F: Inflammatory cytokine and thrombinregulation of interleukin-8 and intercellular adhesion molecule-1expression in first trimester human decidua. J Clin Endocrinol Metab 2005,90:4710-4715.
25. Dodge GR, Poole AR: Immunohistochemical detection andimmunochemical analysis of type II collagen degradation in humannormal, rheumatoid, and osteoarthritic articular cartilages and inexplants of bovine articular cartilage cultured with interleukin 1. J ClinInvest 1989, 83:647-661.
26. van der Kraan PM, van den Berg WB: Anabolic and destructive mediatorsin osteoarthritis. Curr Opin Clin Nutr Metab Care 2000, 3:205-211.
27. Barnes PJ, Karin M: Nuclear factor-kappaB: a pivotal transcription factor inchronic inflammatory diseases. N Engl J Med 1997, 336:1066-1071.
28. van Beijnum JR, Buurman WA, Griffioen AW: Convergence andamplification of toll-like receptor (TLR) and receptor for advancedglycation end products (RAGE) signaling pathways via high mobilitygroup B1 (HMGB1). Angiogenesis 2008, 11:91-99.
29. Pelletier JP, Martel-Pelletier J, Abramson SB: Osteoarthritis, aninflammatory disease: potential implication for the selection of newtherapeutic targets. Arthritis Rheum 2001, 44:1237-1247.
30. Haywood L, McWilliams DF, Pearson CI, Gill SE, Ganesan A, Wilson D,Walsh DA: Inflammation and angiogenesis in osteoarthritis. ArthritisRheum 2003, 48:2173-2177.
31. Fuchs S, Skwara A, Bloch M, Dankbar B: Differential induction andregulation of matrix metalloproteinases in osteoarthritic tissue and fluidsynovial fibroblasts. Osteoarthritis Cartilage 2004, 12:409-418.
32. Bondeson J, Wainwright SD, Lauder S, Amos N, Hughes CE: The role ofsynovial macrophages and macrophage-produced cytokines in drivingaggrecanases, matrix metalloproteinases, and other destructive andinflammatory responses in osteoarthritis. Arthritis Res Ther 2006, 8:R187.
33. Rouhiainen A, Tumova S, Valmu L, Kalkkinen N, Rauvala H: Pivotal advance:analysis of proinflammatory activity of highly purified eukaryoticrecombinant HMGB1 (amphoterin). J Leukoc Biol 2007, 81:49-58.
34. Palumbo R, Galvez BG, Pusterla T, De Marchis F, Cossu G, Marcu KB,Bianchi ME: Cells migrating to sites of tissue damage in response to thedanger signal HMGB1 require NF-kappaB activation. J Cell Biol 2007,179:33-40.
35. Palumbo R, De Marchis F, Pusterla T, Conti A, Alessio M, Bianchi ME: Srcfamily kinases are necessary for cell migration induced by extracellularHMGB1. J Leukoc Biol 2009, 86:617-623.
36. Hreggvidsdottir HS, Ostberg T, Wahamaa H, Schierbeck H, Aveberger AC,Klevenvall L, Palmblad K, Ottosson L, Andersson U, Harris HE: The alarminHMGB1 acts in synergy with endogenous and exogenous danger signalsto promote inflammation. J Leukoc Biol 2009, 86:655-662.
37. Hayashida K, Nanki T, Girschick H, Yavuz S, Ochi T, Lipsky PE: Synovialstromal cells from rheumatoid arthritis patients attract monocytes byproducing MCP-1 and IL-8. Arthritis Res 2001, 3:118-126.
38. Mor A, Abramson SB, Pillinger MH: The fibroblast-like synovial cell inrheumatoid arthritis: a key player in inflammation and joint destruction.Clin Immunol 2005, 115:118-128.
39. Hatano Y, Kasama T, Iwabuchi H, Hanaoka R, Takeuchi HT, Jing L, Mori Y,Kobayashi K, Negishi M, Ide H, Adachi M: Macrophage inflammatoryprotein 1 alpha expression by synovial fluid neutrophils in rheumatoidarthritis. Ann Rheum Dis 1999, 58:297-302.
40. Akahoshi T, Wada C, Endo H, Hirota K, Hosaka S, Takagishi K, Kondo H,Kashiwazaki S, Matsushima K: Expression of monocyte chemotactic andactivating factor in rheumatoid arthritis. Regulation of its production insynovial cells by interleukin-1 and tumor necrosis factor. Arthritis Rheum1993, 36:762-771.
41. Schutyser E, Struyf S, Van Damme J: The CC chemokine CCL20 and itsreceptor CCR6. Cytokine Growth Factor Rev 2003, 14:409-426.
42. Chabaud M, Page G, Miossec P: Enhancing effect of IL-1, IL-17, and TNF-alpha on macrophage inflammatory protein-3 alpha production inrheumatoid arthritis: regulation by soluble receptors and Th2 cytokines.J Immunol 2001, 167:6015-6020.
43. Koch AE, Kunkel SL, Burrows JC, Evanoff HL, Haines GK, Pope RM,Strieter RM: Synovial tissue macrophage as a source of the chemotacticcytokine IL-8. J Immunol 1991, 147:2187-2195.
44. Garcia-Vicuna R, Gomez-Gaviro MV, Dominguez-Luis MJ, Pec MK, Gonzalez-Alvaro I, Alvaro-Gracia JM, Diaz-Gonzalez F: CC and CXC chemokinereceptors mediate migration, proliferation, and matrix metalloproteinaseproduction by fibroblast-like synoviocytes from rheumatoid arthritispatients. Arthritis Rheum 2004, 50:3866-3877.
45. Ruth JH, Shahrara S, Park CC, Morel JC, Kumar P, Qin S, Koch AE: Role ofmacrophage inflammatory protein-3alpha and its ligand CCR6 inrheumatoid arthritis. Lab Invest 2003, 83:579-588.
46. Poole AR, Nelson F, Dahlberg L, Tchetina E, Kobayashi M, Yasuda T,Laverty S, Squires G, Kojima T, Wu W, Billinghurst RC: Proteolysis of thecollagen fibril in osteoarthritis. Biochem Soc Symp 2003, 70:115-123.
47. Billinghurst RC, Dahlberg L, Ionescu M, Reiner A, Bourne R, Rorabeck C,Mitchell P, Hambor J, Diekmann O, Tschesche H, Chen J, Van Wart H,Poole AR: Enhanced cleavage of type II collagen by collagenases inosteoarthritic articular cartilage. J Clin Invest 1997, 99:1534-1545.
48. Wu W, Billinghurst RC, Pidoux I, Antoniou J, Zukor D, Tanzer M, Poole AR:Sites of collagenase cleavage and denaturation of type II collagen inaging and osteoarthritic articular cartilage and their relationship to thedistribution of matrix metalloproteinase 1 and matrix metalloproteinase13. Arthritis Rheum 2002, 46:2087-2094.
49. Murphy G, Cockett MI, Stephens PE, Smith BJ, Docherty AJ: Stromelysin isan activator of procollagenase. A study with natural and recombinantenzymes. Biochem J 1987, 248:265-268.
50. Amos N, Lauder S, Evans A, Feldmann M, Bondeson J: Adenoviral genetransfer into osteoarthritis synovial cells using the endogenous inhibitorIkappaBalpha reveals that most, but not all, inflammatory anddestructive mediators are NFkappaB dependent. Rheumatology 2006,45:1201-1209.
51. Bondeson J, Lauder S, Wainwright S, Amos N, Evans A, Hughes C,Feldmann M, Caterson B: Adenoviral gene transfer of the endogenousinhibitor IkappaBalpha into human osteoarthritis synovial fibroblastsdemonstrates that several matrix metalloproteinases and aggrecanasesare nuclear factor-kappaB-dependent. J Rheumatol 2007, 34:523-533.
52. Miyazawa K, Mori A, Yamamoto K, Okudaira H: Transcriptional roles ofCCAAT/enhancer binding protein-beta, nuclear factor-kappaB, and C-promoter binding factor 1 in interleukin (IL)-1beta-induced IL-6synthesis by human rheumatoid fibroblast-like synoviocytes. J Biol Chem1998, 273:7620-7627.
53. Stein B, Baldwin AS Jr: Distinct mechanisms for regulation of theinterleukin-8 gene involve synergism and cooperativity between C/EBPand NF-kappa B. Mol Cell Biol 1993, 13:7191-7198.
54. Xing L, Remick DG: Promoter elements responsible for antioxidantregulation of MCP-1 gene expression. Antioxid Redox Signal 2007,9:1979-1989.
55. Vincenti MP, Coon CI, Brinckerhoff CE: Nuclear factor kappaB/p50 activatesan element in the distal matrix metalloproteinase 1 promoter ininterleukin-1beta-stimulated synovial fibroblasts. Arthritis Rheum 1998,41:1987-1994.
56. Barchowsky A, Frleta D, Vincenti MP: Integration of the NF-kappaB andmitogen-activated protein kinase/AP-1 pathways at the collagenase-1promoter: divergence of IL-1 and TNF-dependent signal transduction inrabbit primary synovial fibroblasts. Cytokine 2000, 12:1469-1479.
57. Firestein GS, Manning AM: Signal transduction and transcription factors inrheumatic disease. Arthritis Rheum 1999, 42:609-621.
58. Miyazawa K, Mori A, Miyata H, Akahane M, Ajisawa Y, Okudaira H:Regulation of interleukin-1beta-induced interleukin-6 gene expression inhuman fibroblast-like synoviocytes by p38 mitogen-activated proteinkinase. J Biol Chem 1998, 273:24832-24838.
59. Suzuki M, Tetsuka T, Yoshida S, Watanabe N, Kobayashi M, Matsui N,Okamoto T: The role of p38 mitogen-activated protein kinase in IL-6 andIL-8 production from the TNF-alpha- or IL-1beta-stimulated rheumatoidsynovial fibroblasts. FEBS Lett 2000, 465:23-27.
61. Karin M: The regulation of AP-1 activity by mitogen-activated proteinkinases. J Biol Chem 1995, 270:16483-16486.
62. O’Hagan RC, Tozer RG, Symons M, McCormick F, Hassell JA: The activity ofthe Ets transcription factor PEA3 is regulated by two distinct MAPKcascades. Oncogene 1996, 13:1323-1333.
63. Zhang HG, Wang Y, Xie JF, Liang X, Liu D, Yang P, Hsu HC, Ray RB,Mountz JD: Regulation of tumor necrosis factor alpha-mediatedapoptosis of rheumatoid arthritis synovial fibroblasts by the proteinkinase Akt. Arthritis Rheum 2001, 44:1555-1567.
64. El Mabrouk M, Sylvester J, Zafarullah M: Signaling pathways implicated inoncostatin M-induced aggrecanase-1 and matrix metalloproteinase-13expression in human articular chondrocytes. Biochim Biophys Acta 2007,1773:309-320.
65. Litherland GJ, Dixon C, Lakey RL, Robson T, Jones D, Young DA,Cawston TE, Rowan AD: Synergistic collagenase expression and cartilagecollagenolysis are phosphatidylinositol 3-kinase/Akt signaling-dependent. J Biol Chem 2008, 283:14221-14229.
doi:10.1186/ar3124Cite this article as: García-Arnandis et al.: High mobility group box 1potentiates the pro-inflammatory effects of interleukin-1b inosteoarthritic synoviocytes. Arthritis Research & Therapy 2010 12:R165.
Submit your next manuscript to BioMed Centraland take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at www.biomedcentral.com/submit
García-Arnandis et al. Arthritis Research & Therapy 2010, 12:R165http://arthritis-research.com/content/12/4/R165
